Optimal conditions for genetic transformation of the cyanobacterium Anacystis nidulans R2

Under optimal conditions, the cyanobacterium Anacystis nidulans R2 was transformed to ampicillin resistance at frequencies of greater than 10(7) transformants per microgram of plasmid (pCH1) donor DNA. No stringent period of competency was detected, and high frequencies of transformation were achieved with cultures at various growth stages. Transformation increased with time after addition of donor DNA up to 15 to 18 h. The peak of transformation efficiency (transformants/donor molecule) occurred at plasmid concentrations of 125 to 325 ng/ml with an ampicillin resistance donor plasmid (pCH1) and 300 to 625 ng/ml for chloramphenicol resistance conferred by plasmid pSG111. The efficiency of transformation was enhanced by excluding light during the incubation or by blocking photosynthesis with the electron transport inhibitor 3-(3, 4-dichlorophenyl)-1, 1-dimethylurea (DCMU) or the uncoupler carbonyl cyanide-m-chlorophenyl hydrazone. Preincubation of cells in darkness for 15 to 18 h before addition of donor DNA significantly decreased transformation efficiency. Growth of cells in iron-deficient medium before transformation enhanced efficiency fourfold. These results were obtained with selection for ampicillin (pCH1 donor plasmid)- or chloramphenicol (pSG111 donor plasmid)-resistant transformants. Approximately 1,000 transformants per microgram were obtained when chromosomal DNA from an herbicide (DCMU)-resistant mutant was used as donor DNA. DCMU resistance was also transferred to recipient cells by using restriction fragments of chromosomal DNA from DCMU-resistant mutants. This procedure allowed size classes of fragments to be assayed for the presence of the DCMU resistance gene.     

A mutational assay system for L5178Y mouse lymphoma cells, using hypoxanthine-guanine-phosphoribosyl-transferase (HGPRT) -deficiency as marker. The occurrence of a long expression time for mutations induced by X-rays and EMS.

The development of a system for the detection of somatic cell mutation to hypoxanthine-guanine-phosphoribosyl-transferase (HGPRT) (EC 2.4.2.8) deficiency in L5178Y mouse lymphoma cells is described. The selection of mutant cells was not influenced by the concentration of the selective agent 6-thioguanine (6-TG). In addition, all the mutants selected, spontaneous as well as induced ones, showed a complete loss of HGPRT activity. In reconstruction experiments, in which mutant cells were mixed with wild-type cells, the recovery of mutant cells was only markedly influenced when wild-type cells were seeded in a cell density ten times higher than the one, 5-10(4) cells/ml, used in subsequent induction experiments. X-irradiation and treatment with ethyl methanesulfonate (EMS) increased in the mutation rate above the spontaneous background. A clear-cut dose-dependent mutagenic effect after exposure to X-rays was measured. The rate of induced mutations at the HGPRT locus in lymphoma cells was 1-3-10(-7) per R, as determined after exposures of 200, 300, 400, 500 and 600 R. The time the cells needed to express their mutations was much longer than 48 h. Further study of this phenomenon showed that the optimal expression time for TGr-resistant mutants in L5178Y cells was 6 to 7 days. No indication for a dose-dependent effect on the optimal expression of the mutants was found.     

Plasmid mediated mutagenesis of a cellular gene in transfected eukaryotic cells.

NIH3T3 cells are widely used in transformation assays and readily take up transfected DNA. A system has been devised using NIH3T3 cells to measure the mutagenic effect of transfected DNA on recipient cell genes. NIH3T3 cells can be mutated to 6-thioguanine resistance at a frequency which suggests that at least a portion of the cells have only one functional copy of the HGPRT gene. They have a low spontaneous background mutation frequency (approximately 1 X 10(-7)). Transfection of three different plasmids into NIH3T3 cells induced 6-thioguanine resistant mutants at frequencies ranging from 3 to 11 fold above background. The mutant phenotype is stable and reversion frequencies of several mutants are less than or equal to 1 X 10(-7). Southern blot analysis of the HGPRT gene in several mutants showed that 4 of 26 mutants (15.4%) had detectable alterations in the structure of the HGPRT gene. Interestingly 3 of the 4 mutants showing rearrangements were obtained by transfection of the HSV-2 morphological transforming region.     

Quantitative mutagenesis at the adenosine kinase locus in Chinese hamster ovary cells. Development and characteristics of the selection system

Stable mutants exhibiting high degree of resistance (100-1000-fold) to various nucleoside analogs viz, toyocamycin, tubercidin, 6-methyl mercaptopurine riboside (6-MeMPR) and pyrazofurin, are obtained at similar frequency (congruent to 1 X 10(-4] in CHO cells. The mutants resistant to any of the above analogs exhibit similar degree of cross-resistance to the other three nucleoside analogs, and all of the mutants examined contained no measurable activity of the purine salvage pathway enzyme adenosine kinase (AK) which converts these analogs to their phosphorylated derivatives. These results indicate that very similar mutants are selected using any of these analogs. The recovery of AK- mutants in CHO cells is not affected by cell density (up to at least 5 X 10(5) cells per 100-mm diameter dish) and after treatment with mutagen(s) maximum mutagenic effect is observed after 7-8 days, which then remains unchanged for the next several days. Treatment of CHO cells with a number of mutagenic agents e.g. ethyl methanesulfonate, ICR170, ultraviolet light, and benzo[a]pyrene, led to a nearly linear concentration-dependent increase in the frequency of the AK- mutants in cultures. The mutagenic response of the AK locus to these agents compared favorably with that of the HGPRT locus (6-thioguanine resistance) within the same experiments. These results show that the selection system for AK- mutants provides an additional valuable genetic marker for quantitative mutagenesis studies in CHO cells.     

Mutagenicity of an antitumor protein, neocarzinostatin, in Escherichia coli.

An assay is described for the measurement of mutation induction at the hypoxanthine-guanine phosphoribosyl transferase (HGPRT) locus in Chinese hamster ovary (CHO) cells utilizing resistance to 6-thioguanine (TG). Optimal selection conditions are defined for such parameters as phenotypic expression time prior to selection, and TG concentration and cell density which permits maximum mutant recovery. The nature of the TG-resistant mutants is characterized by several physiological and biochemical methods. The data demonstrate that more than 98% of the mutant clones isolated by this selection procedure contain altered HGPRTase activity. The CHO/HGPRT system thus shows the specificity necessary for a specific gene locus mutational assay.     

Induction of 6-thioguanine resistance in human cells treated with N-acetoxy-2-acetylaminofluorene

Induction of 6-thioguanine resistance was studied in human cells treated with the direct-acting chemical carcinogen N-acetoxy-2-acetylaminofluorene (NA-AAF). At low concentrations (2.5–7.5 μM) induction of resistant clones was linear and followed one-hit kinetics, while at 10 μM the yield of resistant clones was higher and appeared to result from the combination of one-hit and two-hit kinetics. A study of about 50 resistant clones revealed that most had reduced levels of hypoxanthine-guanine phosphoribosyl transferase (HGPRT) activity (25–85% of controls) and were able to use exogenous hypoxanthine for growth (“Type II mutants,” deMars, 1974); a few had very low HGPRT activity (1–8% of controls) and were unable to use exogenous hypoxanthine (“Type I mutants”). Use of [9-¹⁴C]NA-AAF allowed us to examine the frequency of induction of thioguanine resistance as a function of binding to DNA (μmole AAF/mole DNA-P). Calculations from these data suggest that most “hits” on the HGPRT locus do not result in detectable mutations: At three different levels of binding and induced mutation frequency, the yield was 2.5-3 detectable mutants/10 000 molecules of acetylaminofluorene bound to the HGPRT locus. These data suggest that most bound acetylaminofluorene molecules either produce no change in the primary sequence of DNA (possibly as a result of repair or correct “read through” by the DNA polymerase) or result in changes which are phenotypically undetectable.     

Cell contact during expression of radiation mutation in Chinese hamster V79 spheroids

The influence of cell to cell contact during expression of radiation mutation at the HGPRT locus was examined using Chinese hamster V79 spheroids. Spheroids left intact for up to 6 days following 7.5 Gy (and then dissociated into single cells for selection in 6-thioguanine) showed no significant decrease in radiation-induced mutation frequency compared to cells of spheroids dissociated immediately following irradiation and passaged in monolayers during the expression interval. These results suggest that the intimate cell contact which occurs between cells in spheroids does not inhibit mutant expression. However, the cell selection process did appear to reduce mutation frequency when spheroids were left intact for 8 days of expression, or when spheroids received 10 Gy.     

Induction of 6-thioguanine resistance in human cells treated with N-acetoxy-2-acetylaminofluorene

Induction of 6-thioguanine resistance was studied in human cells treated with the direct-acting chemical carcinogen N-acetoxy-2-acetylaminofluorene (NA-AAF). At low concentrations (2.5–7.5 μM) induction of resistant clones was linear and followed one-hit kinetics, while at 10 μM the yield of resistant clones was higher and appeared to result from the combination of one-hit and two-hit kinetics. A study of about 50 resistant clones revealed that most had reduced levels of hypoxanthine-guanine phosphoribosyl transferase (HGPRT) activity (25–85% of controls) and were able to use exogenous hypoxanthine for growth (“Type II mutants,” deMars, 1974); a few had very low HGPRT activity (1–8% of controls) and were unable to use exogenous hypoxanthine (“Type I mutants”). Use of [9¹⁴-C]NA-AAF allowed us to examine the frequency of induction of thioguanine resistance as a function of binding to DNA (μmole AAF/mole DNA-P). Calculations from these data suggest that most “hits” on the HGPRT locus do not result in detectable mutations: At three different levels of binding and induced mutation frequency, the yield was 2.5–3 detectable mutants/10 000 molecules of acetylaminofluorene bound to the HGPRT locus. These data suggest that most bound acetylaminofluorene molecules either produce no change in the primary sequence of DNA (possibly as a result of repair or correct “read through” by the DNA polymerase) or result in changes which are phenotypically undetectable.     

Genetic transformation of rifampicin resistance in Lactobacillus acidophilus

Lactobacillus acidophilus strain 100-33, originally isolated from swine faeces, was transformed to rifampicin resistance with DNA from spontaneous rifampicin-resistant mutants derived from it. Cells of the recipient strain were treated with lysozyme and mutanolysin, mixed with donor DNA and polyethylene glycol and grown on a regeneration medium overnight. After 48 h incubation, the numbers of rifampicin-resistant cells in the populations of regenerated cells were estimated from numbers of colonies. Efficiency of the lysozyme/mutanolysin treatment (the ratio of the number of osmotically fragile cells after the enzyme treatment to the initial cell number) was about 99%. The regeneration frequency of the enzyme-treated cells varied from 5 to 67%. The transformation frequency varied from about 0.2 X 10(-8) to 8.0 X 10(-8) transformants per regenerated cell per microgram DNA. To our knowledge, this method for genetic transformation is the first to be reported for a Lactobacillus strain.     

Cellular alterations dependent upon the polyoma virus Hr-t function: separation of mitogenic from transforming capacities.

Hr-t mutants of polyoma virus are restricted in their growth properties (host range) and defective in cell transformation and tumor induction. The present study indicates that these mutants have lost the ability to induce morphological transformation, but have retained a mitogenic function. Thus an early and dramatic difference between wild-type virus and hr-t mutant-infected cultures of rat fibroblasts is the morphological change in individual cells observed by light, fluorescence and scanning electron microscopy. Viruses containing an intact hr-t function (wild-type virus and ts-a mutants) induce a transformed phenotype consisting of stellate cell shape, loss of defined cytoplasmic actin architecture, cellular "underlapping," and increased nuclear and nucleolar sizes. These prominent alterations constitute an abortive transformation, peaking 24-48 hr post-infection, and subsequently resolving in most or all of the cells. In contrast, cells infected with hr-t mutants do not develop the above structural changes, but rather retain their preinfection appearance. Both wild-type virus and hr-t mutants induce cellular DNA synthesis in confluent monolayers of rat cells beginning 12-14 hr post-infection. Flow microfluorometric (FMF) analysis confirms the viral mediated transit of cells from the G1 to the S and G2 phases of the cell cycle, as well as an increase in the proportion of cells with an 8N (octaploid) DNA content. Approximately 50% of the clones isolated from wild-type-infected cultures are polyploid. Stable transformants are found among these polyploid clones, but the majority of the latter resemble the parental cells in their morphology and growth properties. Polyploid clones are derived from hr-t mutant-infected cultures at a much lower frequency, similar to that of mock-infected cultures. Data obtained by sequential labeling of infected cultures with 3 H-thymidine and 5-bromo-deoxyuridine, together with cell number quantitation, indicate that hr-t mutants promote only a single round of cell division, while the wild-type virus and ts-a mutants promote multiple rounds. Loss of the hr-t function in polyoma virus therefore reveals a residual viral mitogenic activity, but prevents the virus from effecting morphological transformation of cells with concomitant loss of defined actin cables, polyploidization and multiple cycles of cell division in confluent cultures.     

HGPRT mutation induction by N-ethyl-N-nitrosourea as measured by 6-thioguanine resistance is higher in male than in female Syrian hamster fetuses

BACKGROUND: The consequences of mutations in embryonic and fetal cells are serious and contribute to high prenatal sensitivity to mutagenic agents. An understanding of the factors that influence the yield of such mutations is important for management of adverse effects of perinatal exposures. Resistance to 6-thioguanine (6-TG) can be utilized to study mutational events at the hypoxanthine-guanine phosphoribosyl transferase (HGPRT) locus. HGPRT is X-linked and recessive. According to the Lyon hypothesis, male cells have only one X-chromosome and female cells randomly inactivate the second X-chromosome. This leads to the prediction that X-linked genes should be equally sensitive to the mutagenic effects of toxicants in male and female fetuses. METHODS: We tested this supposition by in utero exposure of Syrian hamster fetuses to N-ethyl-N-nitrosourea (ENU) at day 12 of gestation. ENU is a strong carcinogen and mutagen. HGPRT mutations were detected by selection with 6-TG. RESULTS: Surprisingly. the male cells had 4 to 5 times more 6-TG mutants than female cells, in two separate experiments (p<0.001). Ouabain resistance, reflecting a co-dominant autosomal locus, was used as a control, and we found that there was no significant difference between male and female cells (p=0.549). CONCLUSIONS: Possible reasons for the sex difference in mutations include escape of the second X-chromosome from inactivation in some of the female cells, or higher mutability in male cells. In any event, there is a gender difference in vulnerability to mutation of an X-linked gene that has previously not been appreciated, and that may be relevant to toxicological studies of such genes. HGPRT is frequently used to monitor mutagenic events in human fetuses.     

Genetic complementation in hybrid cells derived from mutagen-induced mouse clones deficient in HGPRT activity.

Cultured mouse clonal cells, H-5, were treated with two different mutagens, ethyl methanesulfonate (EMS) and N-methyl-N′-nitro-N-nitrosoguanidine (MNNG). Then two selective procedures using 8-azaguanine (8-AZ) or 6-thioguanine (6-TG) were compared in an effort to isolate hypoxanthine-guanine phosphoribosyl-transferase (HGPRT)-deficient cells containing different gene alterations. While many 8-AZ resistant cells were induced by EMS treatment, considerably more 6-TG resistant cells were induced by the same treatment. MNNG treatment induced many 8-AZ resistant mutants but induced hardly any 6-TG resistant mutants. After a fusion experiment of 91 sets involving 13 HGPRT-deficient mouse clones, 7 of which were resistant to 8 AZ and 6 of which were resistant to 6TG with subsequent selection on HAT medium, complementation occurred only in those hybrid mixtures formed between 8-AZ- and 6-TG-resistant clones, while it did not occur at all in hybrid mixtures formed between different 8-AZ-resistant clones and mixtures formed between different 6-TG-resistant clones. The clonally isolated HGPRT-positive cells, characterized by tetraploid karyology, had an apparent activity of HGPRT ranging from 25 to 30% of that of the wild-type parental cells. Heat-inactivation of HGPRT at 65 °C revealed that HGPRT from wild-type cells was heat stable and HGPRT from some 8-AZ-resistant clones were heat labile, while HGPRT from hybrid cells had intermediate stability. These results indicate that there would be alterations in the structural gene of HGPRT in the 8-AZ- or 6-TG-resistant mutants, and also that two selective procedures with 8-AZ or 6-TG alone can isolate different alterations in the structural gene of HGPRT. Moreover, this indicates that some of these gene alterations were mutually complementary. It is most likely that there would be at least 3 cistrons in the locus responsible for HGPRT activity in the mouse cells.     

Agrobacterium-mediated transformation of the filamentous fungus Aspergillus awamori

Many transformation methods have been developed to introduce DNA into filamentous fungi. One of these methods is Agrobacterium-mediated transformation (AMT). Here, we describe an efficient protocol for AMT of Aspergillus awamori. This protocol has been used to determine the function of Agrobacterium virulence genes during AMT, to identify factors influencing transformation frequencies, to generate insertional mutants and to generate A. awamori gene knockout transformants. This protocol in not only applicable to A. awamori, but can be used as a more general guideline for AMT of other filamentous fungi. Conidiospores are incubated with induced Agrobacterium, and, after a cocultivation and selection period, hygromycin-resistant transformants are obtained with a frequency of 200-250 transformants per 1 x 10(6) conidiospores. Using this protocol, transformants can be obtained within 10-12 d.     

Effect of liposomal muramyl tripeptide phosphatidylethanolamine and indomethacin on hematopoietic recovery in irradiated mice

The effects of liposomal muramyl tripeptide phosphatidylethanolamine (MTP-PE/MLV, radioprotective immunomodulator; 10 mg/kg) and indomethacin (INDO, inhibitor of prostaglandin production; 2 mg/kg) on post-irradiation recovery of hematopoietic functions in mice were investigated. Two agents with distinct radioprotective mechanisms were administered alone or in combination 24 h and 3 h before exposure to 7 Gy (60)Co radiation. In the post-irradiation period (3-14 days) combined pre-treatment of mice accelerated recovery of bone marrow cellularity, weight of spleen and myelopoietic and erythropoietic activity in both hematopoietic organs, compared to treatment with MTP-PE/MLV or indomethacin alone. In the peripheral blood, improved radioprotective effects of combined drug administration were found in the recovery of reticulocytes and platelet count. No further significant differences in the recovery of leukocyte count were observed in the examined groups until post-irradiation day 14. Within the first 3-6 post-irradiation days, the bone marrow and peripheral blood smears of mice pre-treated with indomethacin alone or its combination with MTP-PE/MLV more frequently featured blast cells and large cells with abundant cytoplasm which could be considered the hematopoietic stem cells.     

Time versus replication dependence of EMS-induced delayed mutation in Chinese hamster cells

We have previously observed in Chinese hamster cells that ethyl methane sulfonate (EMS) induces mutations which are distributed over at least 10-14 cell divisions following treatment. This delayed appearance of mutations could be explained by EMS-induced lesions which remain in DNA and have a probability that is significantly less than 1.0 of producing base mispairing errors during successive replication cycles (replication-dependent). Alternatively, delayed mutation may be a time-dependent process in which a slow acting or damage inducible error-prone repair process removes persistent DNA lesions and replaces them with an incorrect base during the course of 7-10 days of colony growth following EMS exposure. To address this question, the distribution of HGPRT delayed mutation events (fifth division or later) in cells plated immediately for exponential growth after EMS treatment was compared with the distribution in cells which remained under confluent growth conditions for 8 days and then were replated. Both the distribution and rate of accumulation of delayed mutations (mutations/cell division) were similar in the two culture conditions. In contrast, the frequency of early mutations (before the fifth division) in the confluent population was reduced more than 2-fold compared to dividing cells. A comparison of the frequency of EMS-induced DNA lesions in the two populations revealed that the density inhibited population contained one third the DNA lesions of the exponential population. These results argue against a time-dependent process since, if this mechanism applies, one would expect an increase in early mutant events and a decrease in delayed events in the confluent population. The results, however, are consistent with a replication model in which potential early mutant lesions are preferentially removed in the density inhibited culture during the 8 days of incubation while lesions producing late mutants are not removed.     

Factors affecting the growth of Chinese hamster cells in halt selection media

Factors affecting the efficiency of selection of “reverants” of salvage pathway mutants in media containing amethopterin have been examined. Our V79 Chines hamster cell line was found to require a significantly higher level of thymidine for optimal growth in such media than has been reported for other cell lines. Hypoxanthine (but not glycine) was also required for reversal of amethopterin toxicity, but levels did not differ significantly from those reported elsewhere. Growth in HAT was also dependent on plating density and serum batch. Our modification (VHAT) was compared with published HAT recipies in back selection reconstruction experiments. A sharp fall in EOR (efficiency of recovery) of wild type cells from mixtures with mutants at plating densities greater than 3500 cells/cm² (10⁵ cells/6 cm dish) was observed for VHAT. EOR with other HAT recipes was lower still, and was affected also by the particular mutant used in the mixture.EMS induced “revertants” were isolated from three 8AZ^r mutants by plating in VHAT. All. revertants were however amethopterin resistant, they were also 8AZ resistant and the mobility of residual HGPRT (as measured by polyacrylamide gel electrophoresis) was similar to that of their 8AZ^r parents i.e. dissimilar from that in wild type. The modal chromosome number of V79 wild type cells was 21. No significant deviation from this mode was detected in any of the mutant lines examined. The data indicate that the recovery of colonies in HAT from 8AZ^r mutants does not necessarily indicate that a back mutation in the structural gene for HGPRT has occurred. Thus, the frequency of HAT⁺ colonies cannot be taken as a direct indication of reversion frequencies.     

Agrobacterium-mediated insertional mutagenesis of the ochratoxigenic fungus Aspergillus westerdijkiae

Aspergillus westerdijkiae is a potent ochratoxin A (OTA) producer that has been found in coffee beans. OTA is known to have nephrotoxic effects and carcinogenic potential in animal species. Here we report for the first time the Agrobacterium-mediated transformation for Aspergillus westerdijkiae and the generation of ochratoxin-defective mutants. Conidia were transformed to hygromycin B resistance using strain AGL-1 of Agrobacterium tumefaciens. The obtained transformation frequency was up to 47 transformants per 10(6) target conidia. Among 600 transformants, approximately 5% showed morphological variations. Eight transformants with consistently reduced OTA production were obtained. Two of these transformants did not produce OTA (detection limit: 0.1 microg/kg); the other six mutants produced lower amounts of OTA (1%-32%) compared with the wild-type strain. By using thermal asymmetric interlaced polymerase chain reaction, we successfully identified a putative flavin adenine dinucleotide monooxygenase gene.     

Transformation and mutagenic potential of porphyrin photodynamic therapy in mammalian cells

The transformation and mutagenic potential of porphyrin photodynamic therapy has been examined in mammalian cells. The mutagenic frequency in Chinese hamster cells at the Na+/K+ ATPase locus was measured by resistance to ouabain following treatment with either photodynamic therapy (PDT) or UV irradiation. The C3H 10T 1/2 mouse embryo cell system was used to document the transformation frequency following PDT, UV irradiation, gamma irradiation or exposure to 3-methylcholanthrene (MCA). Treatments with UV irradiation were effective in producing mutants resistant to ouabain, and treatments with UV irradiation, gamma irradiation and MCA generated transformants at frequencies comparable to those which are reported in the literature. However, PDT treatment conditions (which produced a full range of cytotoxicity) did not induce any mutagenic or transformation activity above background levels.     

Genetic instability at the adenine phosphoribosyltransferase locus in mouse L cells.

Resistance to adenine analogs such as 2,6-diaminopurine occurs at a rate of approximately 10(-3) per cell per generation in mouse L cells. This resistance is associated with a loss of detectable adenine phosphoribosyltransferase activity. Other genetic loci in L cells have the expected mutation frequency (approximately 10(-6)). Transformation of L cell mutants with Chinese hamster ovary cell DNA results in transformants with adenine phosphoribosyltransferase activity characteristic of Chinese hamster ovary cells. No activation of the mouse gene occurs on hybridization with human fibroblasts. That this high frequency event is the result of mutation rather than an epigenetic event is supported by antigenic and reversion studies of the 2,6-diaminopurine-resistant clones. These results are consistent with either a mutational hot-spot, a locus specific mutator gene, or a site of integration of an insertion sequence.     

Long-term, low-dose benzo[a]pyrene-induced mutation in human lymphoblasts competent in xenobiotic metabolism

We have studied the mutagenic effects of benzo[a]pyrene (BP) administered in a long-term, low-dose fashion to metabolically competent human lymphoblastoid cells. A continuous dose as low as 0.02 microM for 20 days produced a significant increase in mutant fraction at the 6TG-resistance (HGPRT) locus. The slope of the mutant fraction over time in the 0.02 microM BP-treated culture was twice that observed in the untreated concurrent control; 0.02 microM therefore represents the doubling dose of BP for gene mutation in this cell line. For higher doses of 0.1, 0.5 or 1 microM BP, the rate (or efficiency) of induced mutation was considerably higher for the first 5 or 6 days of exposure than for the last 14-15. This did not appear to be due to a growth disadvantage against early-arising mutants. Comparison to previously published data in the same cell system (Crespi and Thilly, 1984) revealed that the long-term , low-dose protocol (0-1 microM for up to 20 days) was significantly more efficient at inducing mutations than a short-term, high-dose protocol (0-10 microM for 1 day).